Putative cause of vanishing frog populations

Jack Frenkel
November 98

In the Newsletter No. 13 of the American Dendrobatid Group, I summarised an article from the NY Times (16 Sept 1997) about frog declines in Panama and Costa Rica.  In the meantime two articles appeared: Decline of a Tropical Montane Amphibian Fauna, by Dr. Karen R. Lips (Conservation Biology 12 (1), 106-117, February, 1998); the other Chytridiomycosis Causes Amphibian Mortality Associated with Population Declines in the Rain Forests of Australia and Central America, by Dr. Lee Berger and 13 other authors (Proc.Natl. Acad. Sci. USA 95, 9031-9036, July 1998).

Dr. Lips had been conducting surveys of the frog fauna in the mountains of Costa Rica since 1991 and since 1994 had observed a marked reduction in the abundance of several frogs, including Atelopus, Bufo, Hyla and Rana, all of which have both aquatic eggs and larvae.  Those with direct development, such as Eleutherodactylus, were apparently unaffected.  She discusses possible causes for these declines including habitat destruction, introduced rainbow trout, researcher disturbance, ultraviolet irradiation, habitat acidification, altered rainfall patterns, environmental contamination with endocrine-disrupting chemicals as well as a pathogen outbreak.  Furthermore, she compares the Central American decline with other amphibian fauna declines in Australia, Brazil and in Monteverde, Costa Rica (where Bufo periglenus, the Golden frog, had disappeared) starting in 1988 (Crump et al., Copeia, 1992, 413-420).

Dr. Lee and co-authors identified apparently identical Chytrid fungi in sick and dead frogs from Queensland, Australia, (12 species, including Bufo marinus) and Panama (7 species of frogs, including Atelopus, Bufo, Cochranella and Eleutherodactylus, mainly from Fortuna in Northern Panama).  The fungi parasitised the keratinized epidermis, especially of digits and the hypervascularised ventral "drink patch" (presumably interfering with respiration and water absorption and retention).

There was no evidence of viral, bacterial, protozoan or myxozoan infection.  The skin showed little inflammation and the fungus was not cultured.  The skin of archived frogs, collected in previous years, showed no Chytrids.  The fungus was transmitted by skin scrapings from infected Australian frogs.  The animals died between 10 and 18 days after exposure, with fungi in the epidermis; control frogs exposed to filtered (0.45~) skin scrapings remained healthy and were free of Chytrid infection when autopsied after 22 days.  The individual fungus measured about 3x4um with sporangia up to 20um and was cultured in cells but apparently not in artificial media.

Fungi are illustrated in the article from ordinary histological section, from transmission and scanning electron microscopy.  DNA analyses indicated that the amphibian parasite ssu-rDNA (sic) was closely related to another Chytrid fungus.  An outbreak of dying, captive frogs has also been related to Chytrids (J Vet Diagn Investigation, in press, 1998).

Although the infection has not been reported from Dendrobatids or Mantellids, it appears reasonable that they could also be infected.  While the Chytrid infection appears to occur in some localities and not in others, it appears prudent to avoid collecting in Costa Rica and Panama, or at least to conduct a strict quarantine of all introduced frogs.  One should fix dead frogs in 10% formalin and have them examined by an interested pathologist.  It is noteworthy that no examinations of the now extinct Bufo periglenus are available, and recently Dendrobatid frogs were not collected for fear of violating CITES regulations.  Of course CITES permits can be obtained, although often only after overcoming prolonged bureaucratic resistance.